English
Happy New Year!
I provide a series of high-quality AI image generation models that can be used on the Mage platform, including ParagonXL, NovaXL, NovaXL Lightning, NovaXL V2, NovaXL Pony, NovaXL Pony Lightning, RealDreamXL and RealDreamXL Lightning.
Recommended settings for Hyper model:
Sampler: DPM++ 2M Karras or other
Number of steps: 4-6 steps
CFG ratio: 1.5-2.0 (the lower the value, the more variation and lower the contrast)
It is recommended to use ADetailer for generation, and some examples are generated using ADetailer. The models are available at Mage.Space (main sponsor). You can also support me directly on Boosty. All models, including Realistic Vision (VAE/noVAE), are available on Hugging Face.
Reduce image contrast:
You can use LoRA [Detail Tweaker LoRA] and set negative values to reduce image contrast.
Orange: optional
Recommended prompt template:
Positive tip: RAW photo, subject, 8k uhd, DSLR, soft lighting, high quality, film grain, Fuji XT3
Negative hints: (deformed iris, deformed pupil, semi-realistic, cgi, 3d, rendering, sketch, cartoon, painting, animation), text, cropped, out of frame, worst quality, low quality, jpeg Artifacts, ugly, repetitive, sick, mutilated, polydactyly, mutated hands, poorly drawn hands, poorly drawn faces, variation, deformation, blur, dehydration, poor anatomy, wrong proportions, multiple limbs, Cloned face, disfigurement, wrong proportions, misshapen limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, UnrealisticDream (deformed iris, deformed pupil, semi-realistic, cgi, 3d, rendering, sketch, cartoon, painting, anime, mutated hands and fingers: 1.4), (deformed, distorted, disfigured: 1.3), poorly drawn, poor anatomy Structure, Faulty Anatomy, Extra Limb, Missing Limb, Floating Limb, Disconnected Limb, Mutated, Mutated, Ugly, Disgusting, Amputation, UnrealisticDream
Thanks to all model creators for their hard work. Without them, this model would not have been created.
Check whether the network connection is stable, try using a proxy or mirror source; confirm whether you need to log in to your account or provide an API key. If the path or version is wrong, the download will fail.
Make sure you have installed the correct version of the framework, check the version of the dependent libraries required by the model, and update the relevant libraries or switch the supported framework version if necessary.
Use a local cache model to avoid repeated downloads; or switch to a lighter model and optimize the storage path and reading method.
Enable GPU or TPU acceleration, use batch data processing methods, or choose a lightweight model such as MobileNet to increase speed.
Try quantizing the model or using gradient checkpointing to reduce the memory requirements. You can also use distributed computing to spread the task across multiple devices.
Check whether the input data format is correct, whether the preprocessing method matching the model is in place, and if necessary, fine-tune the model to adapt to specific tasks.
